Development and validation of a prognostic model for predicting recurrence in stage I lung adenocarcinoma. With the recent disclosure of results from the National Lung Cancer Screening Trial, suggesting that annual chest CT scans can reduce 20% of lung cancer deaths in high-risk individuals, the number of early stage LACs detected by screening CT scans and their resection is expected to increase. While the 5-year recurrence rate following curative-intent surgical resection in stage I LAC is 18% - 29%, there is substantial individual variation in post-resection recurrence free survival. Currently, the only accepted prognostic factor guiding treatment decisions for both surgeons and oncologists is tumor size, and its prognostic performance remains unclear. Better prognostic tools are needed to provide good quality individual risk prediction, identify patients at high risk of recurrence and help to guide treatment decisions by both oncologists and thoracic surgeons. In an attempt to identify prognostic markers that accurately predict the risk of unfavorable outcome; our group has performed extensive clinical and pathological examination of the largest cohort of stage I LAC to date. We have developed a prediction score that uses combined clinical, histological and cytological criteria that can be assessed routinely on an H&E slide in any hospital, and predicts the risk of recurrence or death with high accuracy. In this proposal, we aim to further improve the predictive ability of our score by microRNA analysis (miRNA). The potential of using microRNAs for prognostication of early lung cancer has been established by several investigators independently including our group. We have developed a microRNA signature capable of prognostication of early lung cancer. A distinct advantage of our methods is the use of formalin-fixed paraffin-embedded (FFPE) tissue, similar to our clinico-pathologic criteria, thus avoiding the requirement of complex tissue processing protocols. We hypothesize that the comprehensive clinico-pathological score, enriched with miRNA expression data, can be combined into a simple, cost-effective predictive instrument that will accurately determine the risk of recurrence or death following curative-intent surgical resection for stage I LAC, and will identify those patients who are primary candidates for aggressive surveillance and adjuvant therapy. In this proposal, we seek to validate and refine our existing miRNA signature for LAC and examine its ability to enrich the predictive quality of our clinico-pathological score. As the model is built upon simple clinical, pathological characteristics (assessed on H&E slide) and miRNA analysis from FFPE, the results of our proposal are immediately implementable, timely and are of high translational significance.